Exponential order statistics and some combinatorial identities
P. Vellaisamy, Aklilu Zeleke
Abstract
P. Vellaisamy, Aklilu Zeleke
Abstract
It is known that the k-th (1≤k≤n) order statistic from the unit exponential distribution can be represented as a sum of independent exponential random variables. We present a proof of this result based on Laplace transform. Also, computing the Laplace transform of the k-th order statistic in two different ways and equating them, we derive several interesting combinatorial identities. A probabilistic interpretation of these identities and their generalizations are also given.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
It is known that the k-th (1≤k≤n) order statistic from the unit exponential distribution can be represented as a sum of independent exponential random variables. We present a proof of this result based on Laplace transform. Also, computing the Laplace transform of the k-th order statistic in two different ways and equating them, we derive several interesting combinatorial identities. A probabilistic interpretation of these identities and their generalizations are also given.
Key concepts: Laplace distribution, Laplace transform, Exponential function, Order statistic, Mathematics, Natural exponential family, Simple (philosophy), Exponential family